From 960179dff90d3d4a8845b72ff923165c1dd9e983 Mon Sep 17 00:00:00 2001
From: helloxteen <helloxteen@foxmail.com>
Date: Thu, 3 Jul 2025 20:02:03 +0800
Subject: [PATCH] modify README.md

---
 kafka/README.md                         | 15 ++++++++++
 load-based-schedule/README.md           | 16 +++++++++-
 omnimv/omnimv-spark-extension/README.md | 39 +++++++------------------
 3 files changed, 40 insertions(+), 30 deletions(-)
 create mode 100644 kafka/README.md

diff --git a/kafka/README.md b/kafka/README.md
new file mode 100644
index 000000000..f174e1969
--- /dev/null
+++ b/kafka/README.md
@@ -0,0 +1,15 @@
+# kafka
+
+#### 介绍
+以Kafka 2.5.0版本使能鲲鹏BoostKit分布式存储算法加速库CRC32加速算法，其他版本可参考本章节使能加速算法。使能该加速算法用于加速Kafka网络传输的CRC32算法计算性能，提升端到端性能。
+
+#### 操作系统
+openEuler 22.03 LTS SP3
+
+#### 使用说明
+用户指南链接：
+https://www.hikunpeng.com/document/detail/zh/kunpengbds/ecosystemEnable/Kafka/kunpengkafkahdp_05_0020.html
+
+#### 参与贡献
+    如果您想为本仓库贡献代码，请向本仓库任意maintainer发送邮件
+    如果您找到产品中的任何Bug，欢迎您提出ISSUE
diff --git a/load-based-schedule/README.md b/load-based-schedule/README.md
index 33b81bc4b..c48c6912e 100644
--- a/load-based-schedule/README.md
+++ b/load-based-schedule/README.md
@@ -1 +1,15 @@
-大数据-Yarn负载调度算法
\ No newline at end of file
+# load-based-schedule
+
+#### 介绍
+Yarn负载调度算法优化开源版本Capacity Scheduler调度器，基于集群节点物理资源权重计算及排序结果进行资源调度，实现资源的均衡配置和高效利用。
+
+#### 操作系统
+openEuler 22.03 LTS SP3
+
+#### 使用说明
+用户指南链接：
+https://www.hikunpeng.com/document/detail/zh/kunpengbds/appAccelFeatures/sqlqueryaccelf/kunpengbds_omniruntime_20_0803.html
+
+#### 参与贡献
+    如果您想为本仓库贡献代码，请向本仓库任意maintainer发送邮件
+    如果您找到产品中的任何Bug，欢迎您提出ISSUE
diff --git a/omnimv/omnimv-spark-extension/README.md b/omnimv/omnimv-spark-extension/README.md
index c4e24f1d0..5f9568ecb 100644
--- a/omnimv/omnimv-spark-extension/README.md
+++ b/omnimv/omnimv-spark-extension/README.md
@@ -1,34 +1,15 @@
 # omnimv-spark-extension
 
-A SQL Engine Extension for Spark SQL to support Materialized View
+#### 介绍
+Spark物化视图插件，提供自动改写SQL的执行计划替换成物化视图的能力。
 
-## Introduction
+#### 操作系统
+openEuler 22.03 LTS SP1
 
-Experiments show that there are duplicate subqueries in batch query,and materialized view is a pre-computing and caching
-technology to reduce repeated calculation in batch query in database field.
-The AI algorithm is used to analyze historical Spark tasks and recommend the optimal materialized view under certain
-conditions.
-The Spark plugin is used to add materialized view management and execution plan rewriting capabilities, greatly
-improving Spark computing efficiency.
+#### 使用说明
+用户指南链接：
+https://www.hikunpeng.com/document/detail/zh/kunpengbds/appAccelFeatures/sqlqueryaccelf/kunpengbds_omniruntime_20_0403.html
 
-## Environment for building OmniMV
-
-```shell
-# download
-wget --no-check-certificate https://mirrors.huaweicloud.com/kunpeng/dist/Apache/hadoop-3.1.1.tar.gz
-# unpack
-tar -zxvf hadoop-3.1.1.tar.gz
-# set environment variables
-export HADOOP_HOME="${pwd}/haddoop-3.1.1"
-```
-
-## Build OmniMV
-
-pull the OmniMV code and compile it to get the jar package
-
-```shell
-git clone https://gitee.com/kunpengcompute/boostkit-bigdata.git
-cd boostkit-bigdata/omnimv/omnimv-spark-extension
-# This step can be compiled, tested and packaged to get plugin/boostkit-omnimv-spark-${omniMV.version}.jar
-mvn clean package
-```
+#### 参与贡献
+    如果您想为本仓库贡献代码，请向本仓库任意maintainer发送邮件
+    如果您找到产品中的任何Bug，欢迎您提出ISSUE
-- 
Gitee